[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <2023082201-armband-unbridle-4893@gregkh>
Date: Tue, 22 Aug 2023 15:34:46 +0200
From: Greg KH <gregkh@...uxfoundation.org>
To: Boris Lukashev <blukashev@...pervictus.com>
Cc: kernel-hardening@...ts.openwall.com
Subject: Re: [PATCH] Restrict access to TIOCLINUX
On Tue, Aug 22, 2023 at 08:51:03AM -0400, Boris Lukashev wrote:
> See this asked a lot, and as someone who works mostly in languages
> with robust test facilities, I gotta ask: why isn't kernel code
> mandated to be submitted with tests?
Because for most/many of the kernel api, we don't have tests yet. LTP
covers a lot of this type of thing, so no need to duplicate that. But
yes, if you want to mandate this for your subsystem, that would be
great!
And many subsystems do mandate that, like drm and bpf. So if you want
to bring the current LTP tty tests into the kernel test framework, I'll
gladly take those patches.
But I fail to see how this is relevant for this proposed change, which
would restrict access to an ioctl, how would a test help here?
thanks,
greg k-h
Powered by blists - more mailing lists
Confused about mailing lists and their use?
Read about mailing lists on Wikipedia
and check out these
guidelines on proper formatting of your messages.
